Auditory deficits in infants at risk for dyslexia during a linguistic sensitive period predict future language

Abstract

Highlights Auditory functioning was assessed in infants at risk for dyslexia using MEG. Atypical larger and longer neural responses in at-risk infants at 12 than 6 months. Responses were estimated to come from left temporal and left prefrontal regions. Neural responses predicted subsequent language skills evaluated at 18–30 months. These responses may serve as an early marker of risk for dyslexia.

Reduced Theta Sampling in Infants at Risk for Dyslexia across the Sensitive Period of Native Phoneme Learning

“…Moreover, because gamma rhythms were linked to phonological processing deficits in dyslexia [20,22], we also hypothesized to find atypical locking at gamma in at-risk infants. Finally, consistent with our position that auditory sampling during this sensitive period is critical to language learning and our prior work showing that infants' early neural responses to simple sounds and language can predict later language [43,44], we further hypothesized that atypical auditory sampling at frequencies relevant for speech processing in at-risk infants would predict functional outcomes of later language skills. To test this, we correlated stimulus locking at frequencies relevant for language processing with later nonlinguistic communication and perceptive, expressive, and syntactic language skills at 13-30 months of age because similar measures were found to predict later literacy skills [45,46].…”

“…When examining the 6-and 12-month-old data separately for each group, we found that this effect likely stems from the 6-month-old infants in the control group, r = −0.566, n = 11, p = 0.069; 12-month-old, p = 0.44 (15 months), and from the 12-month-old infants in the at-risk group, r = −0.473, n = 14, p = 0.087; 6-month-old, p = 0.133 (13 months), r = −0.57, n = 14, p = 0.033; 6-month-old, p = 0.218 (15 months). It is difficult to speculate about the exact cause of these results, but we know that infants' brain mechanisms mature across the sensitive period for native phoneme learning, such as TD infants' brains become more efficient in processing simple white noise between 6 and 12 months [44]. Although this is speculative, our ASSR-CDI results suggest that such efficiency processing seems to happen later in infants at risk for dyslexia compared to their age-matched controls.…”

Improving Localization Accuracy of Neural Sources by Pre-processing: Demonstration With Infant MEG Data

“…Figure 1 shows field topographies of real data from a 6-month-old subject from Mittag et al ( 28 ), averaged across trials with an auditory stimulus, processed in three different ways. In Figure 1A no compensation is done for subject movement, and the magnetic field topography of the evoked response is clearly adversely affected by external artifacts and subject movement.…”

“…Using simulated data distorted by head movements from real recordings, we analyzed localization bias and goodness-of-fit under the four different processing strategies described in Methods: Data Processing. The head movements were drawn from Kuhl et al ( 27 ) (7 mo) & Mittag et al ( 28 ) (6 and 12 mo). The top panel of Figure 2 shows that between the first experiment ( 27 ) and the second experiment ( 28 ), there was an improvement in subject compliance in terms of reduced movement, as reflected in less subject deviation from the mean head position.…”

“…Infants were seated in a custom-made chair under the MEG helmet while listening to various auditory stimuli. For specific details about the paradigm, see Kuhl et al ( 27 ) and Mittag et al ( 28 ).…”