Neural evidence of the allophonic mode of speech perception in adults with dyslexia

The Categorical Perception Deficit in Dyslexia: A Meta-Analysis

“…Even though allophonic sensitivity was no longer apparent in the behavioral responses of the Dutch children at risk for dyslexia in first grade, it was still present in neurophysiological recordings performed on the same children (Noordenbos, Segers, Serniclaes, Mitterer, & Verhoeven, 2012b). Similar results have been obtained for adults with dyslexia (Noordenbos, Segers, Serniclaes, & Verhoeven, 2013). These apparently conflicting results might be explained by differences in sensitivity between behavioral and neural measures.…”

Individual Differences in Subphonemic Sensitivity and Phonological Skills

“…Specifically, the 784 spectrotemporal sensitivity of the superior temporal gyrus (STG) has been linked to sensitivity to 785 phonetic features, such as voice onset time, place of articulation, and formant frequency (for a 786 review, see Leonard & Chang, 2014). Given functional and structural deviations in the STG 787 (Maisog, Einbinder, Flowers, Turkeltaub, & Eden, 2008;Paulesu et al, 2001;Simos et al, 2002;788 Steinbrink et al, 2008) and heightened sensitivity to phonetic features (e.g., Bogliotti et al, 789 2008;Noordenbos et al, 2013Noordenbos et al, , 2012aNoordenbos et al, , 2012bSerniclaes et al, 2004) …”

“…On these accounts, RD individuals' phonological representations are either under-or 30 overspecified (as labelled by Noordenbos, Segers, Serniclaes, & Verhoeven, 2013). The 31 underspecification account suggests that RD individuals' poorer performance on meta-32 phonological tasks originate from incomplete or imprecise encoding of speech.…”

“…For 130 instance, two longitudinal studies carried out in the USA (Molfese, 2000;Molfese & Molfese, 131 1997;Molfese, Molfese, & Modgline, 2001) and Finland (Guttorm et al, 2005;Guttorm, 132 Leppänen, Tolvanen, & Lyytinen, 2003;Lyytinen et al, 2004) provide evidence that differences 133 in event-related potentials (ERPs) in response to speech and non-speech auditory signals at birth 134 (e.g., N1 peak latency, N2 peak amplitude, mean amplitude, mismatch negativity) may predict 135 subsequent differences in oral language and literacy skills in the preschool and early grade 136 school years. Furthermore, individuals at risk for or with RD, whose performance in behavioral 137 categorical speech perception tasks is comparable with that of typical readers, still show neural 138 sensitivity to allophonic contrasts as indexed by the mismatch negativity (MMN) component of 139 ERP (Noordenbos et al, 2012b;Noordenbos et al, 2013). This implies that, despite 140 indistinguishable behavioral judgment in categorical perception, subtle differences of 141 phonological perception between typically developing vs. RD individuals can be detected with 142 more sensitive measures of automatic, online processing.…”

“…Therefore, behavior 96 indicating greater subphonemic sensitivity would be consistent with the overspecification 97 hypothesis and at odds with underspecification. 98 less steep identification slopes, lower peak discrimination at the typical boundary, and additional 121 discrimination peaks at within-category stimulus pairs that often align with phonetic boundaries 122 between allophones (Noordenbos et al, 2012a(Noordenbos et al, , 2013Serniclaes et al, 2001Serniclaes et al, , 2004, suggesting 123 phonological representations organized allophonically rather than phonemically (Serniclaes, 124 2006). Although categorical perception tasks have proved fruitful in assessing underlying 125 phonological representations, they nevertheless require post-perceptual meta-linguistic 126 judgments, and so might not be sensitive to subtleties of online speech processing.…”